Low-Strength Wastewater Treatment with Combined Granular Anaerobic and Suspended Aerobic Cultures in Upflow Sludge Blanket Reactors

Vedat Yılmaz1

G�ksel N. Demirer2

1Department of EnvironmentalEngineering, Akdeniz University,Antalya, Turkey.

2Department of EnvironmentalEngineering, Middle East TechnicalUniversity, Ankara, Turkey.

Research Article

Enhancing the Performance of Anaerobic Digestionof Dairy Manure through Phase-Separation

Anaerobic digestion (AD) is an effective way to convert animal manures into profit-able by-products while simultaneously reducing the pollution of water, air, and soilcaused by these wastes. Conventional high-rate anaerobic reactors cannot effectivelyprocess animal manures with high solids-containing wastes. The two-phase configura-tion for AD has several advantages over conventional one-phase processes, e.g.,increased stability of the process, smaller size and cost efficient process configura-tions. In the present study, the experiments were carried out in a two-phase systemcomposed of an acidogenic reactor and a methanogenic reactor, and in a one-phasesystem composed of only a methanogenic reactor. The reactors were operated asunmixed (without an external mixing aid), unsophisticated, and daily-fed mode. Itwas found that the two-phase configuration was more efficient than the one-phasesystem. The biogas production in the two-phase system at a hydraulic retention time(HRT) of 8.6 days (only methanogenic phase) was calculated to be 42% higher at anorganic loading rate (OLR) of 3.5 g VS/L N day than that of the one-phase with a HRT of20 days. This translates into significant performance improvement and reduced vol-ume requirement. This finding represents a further step in the achievement of wideruse of simple anaerobic reactor configurations for waste treatment in rural areas.

Keywords: Anaerobic digestion; Dairy manure; Methane; Two-phase;

Received: January 30, 2008; revised: April 17, 2008; accepted: May 20, 2008

DOI: 10.1002/clen.200800024

1 Introduction

Manure residues from livestock industries have been identified as amajor source of environmental pollution. The most common prob-lems directly linked to untreated animal manure are odor, methaneand ammonia emissions, and the release of nutrients and patho-gens that may affect human health [1 – 3]. Traditionally, thesewastes have been disposed of, directly or after composting, as soilsupplements in the agricultural industry [4, 5]. Since this practicehas resulted in the degradation of air, soil, and water resources,new regulations for the protection of the environment have beenpromulgated to control land application of animal manure [6].

Anaerobic digestion (AD) of biomass is an alternative not only forwaste minimization but also for renewable energy production [7, 8].This biological process also kills pathogens, reduces odors, and pro-duces a stable solid digestate that can be utilized as a compost orsoil conditioner [9 – 11].

Conventional one-phase slurry AD is not an effective system forwastes containing high levels of solids (A10%), since they requirethe manure that is capable of being pumped, which itself necessi-tates a solids concentration of a10%. This, in turn, results in a signif-icant increase in fluid and digester volume, which causes increased

capital and operating costs [12]. Instability or failure of one-phasemethanogenic digesters has been widely reported for a variety ofwastewaters, especially under high loading conditions [13, 14]. Prob-lems encountered with stability and controls in conventionaldesign applications have led researchers to search for new solutions[15].

The two-phase configuration for AD has several advantages overconventional one-phase processes, e. g., increased stability of theprocess, smaller size and cost efficient process configurations. Onerelevant feature of the two-phase AD is that when a high solids-con-taining waste is introduced to the first phase it is liquefied alongwith acidification. This translates into less liquid addition and,thus, less energy requirements for heating, storing, etc. Eventhough several aspects of two-phase configuration including lique-faction might be very significant for efficient AD of dairy manure,its application on animal manure is very limited. Finally, the config-uration can be smaller in size and more cost effective [16]. Theresults of several studies have clearly demonstrated the applicabil-ity and efficiency of two-phase AD for high solids-containing wastes.

In spite of these advantages, the use of anaerobic digestion fortreating cow manure has not been popular in Turkey. The conven-tional AD of animal manure was attempted in farm-scale digestersat several locations across Turkey during the period 1978 – 1982 andfound to be unsuccessful because of high heating costs, the need forskilled operators and operational instability. This failure mainlyresulted from low sensitivity to the protection of the environmentby farmers, improper process design, and frequent operation fail-

Correspondence: Dr. G. N. Demirer, Department of Environmental Engi-neering, Middle East Technical University, In�n� Bulvari, 06531, An-kara, Turkey.E-mail: [email protected]

i 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.clean-journal.com

760 Clean 2008, 36 (9), 760 –766

Clean 2008, 36 (9), 760 – 766 Anaerobic Digestion of Dairy Manure 761

ures. However, an economic evaluation of these digesters nowreveals that the plants can be economically viable. In addition, astrong demand for renewable energy generation, new and forth-coming environmental legislation, e. g., climate change levies, hasgradually increased the interest in anaerobic digestion technology[17].

This work aims to evaluate the feasibility of a two-phase anaerobictreatment system for unscreened dairy manure. The specific objec-tive is to compare biogas production of the one- and two-phasedigestion processes at operating temperatures of 35 and 258C.

2 Materials and Methods

Wet manure was collected at three different times from a privatedairy in G�lbasi, Ankara, and stored at 48C prior to use. The compo-sition of the dairy manure used in this study had the following char-acteristics: total solids (TS) of 20.1 l 1.7%, volatile solids (VS) of 67 l4.6% of TS and density of 1028 l 20 g/L. The raw manure was dilutedwith water to decrease the solids content to achieve slurries with3.5 and 15 g VS/L. The ratio between chemical oxygen demand(COD) and VS of this manure was found to be 1.04.

The mixed anaerobic culture used as seed was obtained from theanaerobic sludge digesters of the municipal wastewater treatmentplant in Ankara, with a solids retention time (SRT) of 14 days. Themixed anaerobic culture was concentrated by settling before beingused as an inoculum. The volatile suspended solids (VSS) concentra-tion of the concentrated seed cultures used was 23930 l 3162 mg/L.

2.1 Experimental Setup

A schematic representation of the laboratory-scale, one-phase (R1)and two-phase (R21 and R22) anaerobic digestion systems used inthe experimental system is depicted in Fig. 1. The effective volumesof R1, R21, and R22 were 1000, 400, and 1000 mL, respectively. Thetwo-phase configuration contained R21 and R22 as the first (acido-genic) and second (methanogenic) phases. The solids and hydraulicretention times (SRT/HRT) applied to each reactor were similar sinceno recycling of the effluent was performed. The SRT/HRT values ofR1, R21, and R22 were 20, 2, and 8.6 days, respectively. The overalltwo-phase configuration could be taken as a SRT/HRT value of 10.6days. All the reactors were fed daily. The gas production in R1, R21,and R22 were monitored by a water replacement device. One set ofreactors were maintained at 258C in a temperature-controlled waterbath and the others were held at 35 l 28C in a controlled room, andall reactors were shaken manually once daily after conducting thegas production measurement.

R1, R21, and R22 were seeded with 500, 200 and 500 mL of mixedanaerobic seed culture, respectively. Two stock solutions were pre-pared prior to reactor feeding for targeted VS concentrations. Thereactors were operated without any addition of chemicals for pHand alkalinity control, in acidogenic and methanogenic reactors.

The performance of the reactors was monitored by measuring bio-gas production and soluble COD, VS, volatile fatty acid (VFA), pH,and oxidation reduction potential (ORP).

2.2 Analytical Methods

The pH, daily gas production, total solids, volatile solids, methanepercentage, total volatile fatty acids (tVFA) and effluent soluble COD

(sCOD) were monitored in each reactor. pH, TS, and VS analysis wereperformed using Standard Methods [18]. sCOD was measured usingHach COD vials according to the EPA approved digestion method[19]. Accordingly, after 2 h digestion, sample sCOD values weredirectly read using a Hach 45 600-02 spectrophotometer (Hach Co.Loveland, Co., USA).

2.2.1 Total Volatile Fatty Acids

The filtered samples were acidified with 99% formic acid to a pH ofless than 3 to convert the fatty acids to their undissociated forms,i. e., acetic acid, propionic acid, butyric acid, etc., before injecting 1lL of the acidified samples into a gas chromatograph (GC) [20]. Foreach individual VFA analysis, a GC unit (Thermo Electron Co.)equipped with a flame ionization detector fitted with a Zebron ZB-FFAP column, with a length of 30 m, i.d. of 0.32 mm and film thick-ness of 0.25 lm, was used. The concentration of each VFA species,i. e., acetic, propionic, butyric, isobutyric, valeric, isovaleric and cap-roic acids, was expressed in terms of acetic acid (Hac) equivalents bydividing the concentration value by its molecular weight and multi-plying by the molecular weight of acetic acid. The tVFA value wasdetermined by the sum of the concentrations of all VFA speciesexpressed in terms of acetic acid equivalents.

2.2.2 Gas Analysis

The total gas produced in the reactors was measured by connectingthe reactor headspace to a water displacement column filled withdistilled water and recording the volume of displaced solution. Gassamples for gas composition analysis were periodically withdrawnfrom the reactors by a 100 lL Hamilton gas-tight glass syringe. Thegas composition was determined by a GC unit (Thermo Electron Co.)equipped with a thermal conductivity detector. Methane, nitrogenand carbon dioxide were separated through a 15 m Porapak Q, 5mm i.d. column.

3 Results and Discussion

Average methane production values of 775 l 105, 650 l 50, 485 l 45,125 l 23 mL/day were obtained for the reactors R1(35), R22(35),R1(25), R22(25), respectively, see Fig. 2. In addition, a significantamount of methane production, specifically 35 mL, was seen in the


Figure 1. Experimental setup used in the study.

762 V. Yılmaz and G. N. Demirer Clean 2008, 36 (9), 760 –766

mesophilic acidogenic reactor, i. e., R21(35). As can be seen from Fig.2, three different gas production trends were observed. This couldbe explained by the heterogeneous characteristics of the manuresamples collected at different times resulting in different biode-gradability yields.

An increase in the temperature results in an increase in the rateof hydrolysis, acidification and methanogenesis stages under one-phase conditions, see Fig. 2. In the one-phase reactor, the biogas pro-duction increased by 56% on increasing the temperature from 25 to358C. The results obtained are in agreement with previous results,which showed an improvement in biogas yields with increasingtemperature [21, 22]. In the mesophilic temperature range (25 –358C), higher temperatures tend to give better methane yields. Thebiogas production trend is much more stable at 35 than 258C.

The methane yields of R1(35), R22(35), R1(25), and R22(25) werecalculated as 221, 216, 132 and 43 mL CH4/(g VS added), respectively.The performance of the reactors in terms of biogas yield is compara-ble with literature values except for reactor R22(25) [23 – 25]. In themesophilic temperature range, the performance of the two-phasesystem (216 mL CH4/g VS) is slightly lower than the one-phase system(221 mL CH4/g VS). Previous findings with cattle waste demonstratethat upon decreasing the HRT from 20 to 10 days, a considerabledecrease is observed in the amount of biogas produced [16, 21, 26,27]. From these conclusions, a simple analysis could easily revealwhich system is more preferable in terms of biogas productionyield. When the HRT of the two-phase system is increased from 8.6

to 20 days, the system will produce at least 307 mL methane insteadof 216 mL CH4/g VS. Thus, the gas production in R22(35) will be 42%higher than R1(35). Moreover, a small amount of methane producedfrom the acid phase, i. e., R21(35), may also be delivered to R22(35)or directly collected; which will definitely increase the amount ofmethane production in R2.

The methane content of the reactors R1(35), R22(35), and R1(25)almost constitute 63 to 65% of the biogas produced, see Fig. 3. Thesethree reactors produced similar percentages of CO2 and N2. R22(25)did not provide similar results since its biogas production efficiencywas significantly lower than the other reactors. It typically con-tained 40 to 45% CH4, 30 to 35% CO2, and 25 to 30% N2. This low effi-ciency might be as a result of the low operating temperature.

The acidogenic reactors had different biogas compositions andproductions than the methanogenic reactors. The gas production inR21(25) was not considerable and ca. 130 mL gas was produced inR21(35). The reactor R21(35) contained 30 to 35% CH4, 25 to 30%CO2, and 40 to 45% N2, see Fig. 3. In a previous publication [28], themethane production in acidogenic reactors varied between 5 to 27%at different OLRs and SRTs. The methane production at such lowSRTs could be explained by unintended extension of the retentiontimes of microorganisms in the reactors due to very high solids con-centration [29]. The optimum conditions for acidification severelyretard methanogenic activity but do not eliminate all methano-gens, which are sensitive to the operating conditions but may per-sist in a dormant or semi-dormant state. GC analyses also indicatedan unexpected concentration of N2 (80 – 85%) in the reactor R21(25).This high concentration was mainly due to low biogas production(10 – 20 mL) in the reactor which caused unavoidable analyticalerrors in the GC measurements. However, with the support of litera-ture data, 40 to 45% of the nitrogen content of the biogas from thereactor R21(35) can be explained by the fact that denitrificationmight be responsible for simultaneous VFA production and nitrateelimination [30, 31]. Although the importance of the separation ofthe acidogenic and methanogenic phases is well known, only a fewstudies have been carried out for the investigation of the acidogenicphase of AD with manure and they have not focused on the forma-tion of nitrogen.

The influent and effluent VS concentrations in the reactors areplotted in Fig. 4a). The highest VS conversion was observed as 35 to40% in the reactor R1(35) between days 10 and 100, but during thedays 100 to 200 the reactor R2(35) had the highest VS reduction rateof 30 to 35%. Since the operation in the reactor R1(25) was not sta-ble, a wide range of VS conversion was seen in this reactor, i. e., 20 to30%. The volatile solids removal was lower compared to the valuesfound in a similar run at 358C. The VS reduction observed in R2(25)and R21(35) was under 20% compared to their gas production andthe values fluctuated dramatically. Although both systems had thesame OLR relative to their inlet concentrations, the inlet concentra-tion of R22 was the effluent concentration of R21 in which therewas an average VS reduction of 17%. Therefore, the OLR in R22 wascalculated as 2.9 g VS/L N day. The reactor R22(35) had 10 to 50%higher VS reduction than the reactor R22(25), since the perform-ance of R22(25) was low. The VS reduction in R21(25) was almostalways below 10%. The effluent VS concentrations revealed a stabletrend especially in the mesophilic reactors. This stable trendshowed that a constant VS reduction occurred throughout the oper-ation.

The amount of sCOD in the reactors R21(35) and R21(25)increased by 65 and 25%, respectively, as shown in Fig. 4b). The solu-


Figure 2. Daily methane productions at 35 and 258C.


bilization also increased with increasing temperature. The hydroly-sis and solubilization of complex materials is the main mechanismin the acidification phase, and thus, the efficiency of system couldbe easily promoted by increasing the temperature. The sCOD reduc-tions of the reactors R1(35), R22(35), R1(25) were found to be 45, 60and 52%, respectively.

The effluent tVFA concentrations of the first-phase reactor (acido-genic-phase) at mesophilic and low temperature increased from700 – 750 to 1700 – 1880 and 1125– 1280 mg/L (as Hac), respectively,see Fig. 4c). These effluent concentrations revealed more than 100and 60% increase over the influent values for the reactors R21(35)and R21(25), respectively. The effluent VFA concentration of the sec-ond-stage reactor at mesophilic temperature decreased to 350 mg/L(as Hac), but the effluent concentration of the reactor R22(25)

remained constant as expected. The tVFA concentration of the reac-tor R1(25) was much lower than that of R22(25), since the biogas pro-duction in R1(25) was more than double that of R22(25). Thisresulted in more VFA consumption in the reactor R1(25). Although,the acidogenic efficiency could be increased with temperature, mix-ing and pH control were not important parameters for the effi-ciency [28].

The total effluent VFA value in the one-phase reactor was lowerthan that of the two-phase reactor at mesophilic temperature, i. e.,the efficiency of the two-phase system was higher than that of theone-phase system in terms of VFA consumption. However, this doesnot mean that more VFAs were converted to methane in the one-phase reactor, since more VFA transferred from the reactor R21(35)to R22(35).


Figure 3. CH4, CO2, and N2 contents in the reactors (in %).


The pH values of the reactors R21(35), R22(35), R1(35), R21(25),R22(25), and R1(25) were 6.5 – 6.8, 7.0 – 7.2, 7.3 – 7.5, 7.0 – 7.2, 6.8 –7.0, and 7.2 – 7.5, respectively. Cattle manure is a complex substratecontaining undissolved and dissolved organic matter, e. g., polysac-charides, lipids, proteins and inorganic compounds of importancefor the chemical environment. Therefore, pH levels were notexpected to be lower than 6.5 in the acidogenic phase.

The VFA:COD ratio is a measure of the degree of success of acido-genesis, representing the amount of solubilized matter that hasbeen converted to VFAs [32]. As can be seen from Fig. 5, the degree ofacidification increased with increasing temperature and by the useof the two-phase configuration. The one-phase reactors resulted inthe lowest acidification formation with 4 to 5% in reactors R1(35)and R1(25). The reactors R21(35) and R22(35) revealed an almost sta-

ble acidification degree of 30 and 8 to 10%, respectively. The reac-tors R21(25) and R22(25) showed almost the same trend , with theresults being a little lower than R21(35).

4 Conclusions

This study revealed that two-phase AD of high solids containing ani-mal manure can be easily operated without any pH adjustment orrequirement for continuous mixing. Based on the results of thisstudy, the following conclusions were drawn:(i) The daily biogas production in the one-phase reactor was 1300

and 800 mL at mesophilic temperature and at low temperature(258C), respectively.


Figure 4. VS, sCOD, and tVFA concentrations in the reactors.


(ii) The use of a two-phase reactor at a HRT of 10.6 days (two daysacidogenic and 8.6 days methanogenic) for AD of dairy manureresulted in 42% higher biogas production (based on the experi-mental results and literature data) relative to a conventionalone-phase configuration with a HRT of 20 days.

(iii) The degree of acidification is higher at the mesophilic temper-ature (30%) than at low temperature (25%).

(iv) Due to the high N2 content of the biogas from the acidogenicreactors, and with the support of literature data, it can bespeculated that denitrification might be responsible fromsimultaneous VFA production and nitrate elimination.

(v) The tVFA concentrations were reduced from 1700 to 400 mgHac/L and 1300 to 300 mg Hac/L in reactors R22(35) and R1(35),respectively. This implies that the performance of the two-phase system was higher than the one-phase system in terms ofVFA conversion.

Acknowledgment

This study was funded by the Scientific and Technological Councilof Turkey through Grant Number 104I127.

References

[1] A. Wellinger, Anaerobic digestion: A review comparison with twotypes of aeration systems for manure treatment and energy produc-tion on the small farm, Agric. Wastes 1984, 10, 117 – 133.

[2] R. I. Mackie, P. G. Stroot, V. H. Varel, Biochemical identification andbiological origin of key odor components in livestock waste, J. Anim.Sci. 1998, 76, 1331 – 1342.

[3] A. C. Wilkie, Anaerobic digestion: holistic bioprocessing of animalmanures, Proc. of the Animal Residuals Management Conf., Water Envi-ronment Federation, Alexandria, VA 2000, pp. 1 – 12.

[4] H. H. van Horn, A. C. Wilkie, W. J. Powers, R. A. Nordstedt, Compo-nents of dairy waste management systems, J. Dairy Sci. 1994, 77,2008 – 2030.

[5] Environmental monitoring by feedlots, National Animal Health Moni-toring System, USDA, Fort Collins, CO 1995.

[6] Guide manual on NPDES regulations for concentrated animal feed-ing operations, EPA 833-B-95-001, US EPA, Washington, DC 1995.

[7] D. P. Chynoweth, J. M. Owens, R. Legrand, Renewable methane fromanaerobic digestion of biomass, Renewable Energy 2001, 22, 1 – 8.

[8] L. T. Angenent, S. Sung, L. Ruskin, Methanogenic population dynam-ics during startup of a full-scale anaerobic sequencing batch reactortreating swine waste, Water Res. 2002, 36, 4648 – 4654.

[9] A. G. Hashimoto, T. G. Edgar, H. Nakano, Methane production fromanimal wastes, J. Soil Water Conserv. 1994, 49 (2), 62 – 65.

[10] D. P. Chynoweth, Environmental impact of biomethanogenesis, Envi-ron. Monit. Assess. 1996, 42 (1 – 2), 3 – 18.

[11] P. E. Wrigth, S. F. Inglis, S. M. Stefman, J. F. Bonhotal, Reduction ofselected pathogens in anaerobic digestion, Proc. of the Ninth Int. Symp.Animal, Agricultural and Food Processing Wastes. ASAE Pub., Raleigh, NC2003, pp. 74 – 82.

[12] G. N. Demirer, S. Chen, Effect of retention time and organic loadingrate on anaerobic acidification and biogasification of dairy manure,J. Chem. Technol. Biotechnol. 2004, 79 (12), 1381 – 1387.

[13] P. Fox, F. G. Pohland, Anaerobic treatment applications and funda-mentals: substrate specifity during phase separation, Water Environ.Res. 1994, 66 (5), 716 – 723.

[14] S. Ghosh et al., Pilot- and full-scale two-phase anaerobic digestion ofmunicipal sludge, Water Environ. Res. 1995, 67 (2), 206 – 214.

[15] B. Demirel, O. Yenigun, Anaerobic acidogenesis of dairy wastewater:the effects of variations in hydraulic retention time with no pH con-trol, J. Chem. Technol. Biotechnol. 2004, 79 (7), 755 – 760.

[16] G. N. Demirer, S. Chen, Two-phase anaerobic digestion ofunscreened dairy manure, Process Biochem. 2005, 40 (11), 3542 – 3549.

[17] J. B. van Lier et al., New perspectives in anaerobic digestion, WaterSci. Technol. 2001, 43 (1), 1 – 18.

[18] Standard Methods for the Examination of Water and Wastewater, 19th ed.,APHA (American Public Health Association), Washington, DC 1995.

[19] HACH Water Analysis Handbook, 2nd ed., HACH Company, Loveland,CO 1992.

[20] G. Dinopoulou, T. Rudd, J. N. Lester, Anaerobic acidogenesis of acomplex wastewater: I. The influence of operational parameters onreactor performance, Biotechnol. Bioeng. 1988, 31 (6), 958 – 968.

[21] R. Summers, N. Hobson, C. R. Harries, A. J. Richardson, Stirred-tank,mesophilic, anaerobic digestion of fattening-cattle wastes and ofwhole and separated dairy-cattle wastes, Biol. Wastes 1987, 20 (1), 43 –62.

[22] V. H. Varel, A. G. Hashimoto, Y. R. Chen, Effect of temperature andretention time on methane production from beef cattle waste, Appl.Environ. Microbiol. 1980, 40 (2), 217 – 222.

[23] Z. Mladenovska, Z. Dabrowski, B. K. Ahring, Anaerobic digestion ofmanure and mixture of manure with lipids: biogas reactor perform-ance and microbial community analysis, Water Sci. Technol. 2003, 48(6), 271 – 278.


Figure 5. Degree of acidification in the reactors.


[24] R. I. Mackie, M. P. Bryant, Anaerobic digestion of cattle waste at mes-ophilic and thermophilic temperatures, Appl. Microbiol. Biotechnol.1995, 43, 346 – 350.

[25] V. H. Varel, A. G. Hashimoto, Y. R. Chen, Effect of temperature andretention time on methane production from beef cattle waste, Appl.Microbiol. Biotechnol. 1980, 40 (2), 217 – 222.

[26] A. Wellinger, Process design of agricultural digesters. Nova EnergieGmbH, Ettenhausen, Switzerland 1999 (Retrieved April 12, 2007from: http://homepage2.nifty.com/biogas/cnt/refdoc/whrefdoc/d14prdgn.pdf).

[27] P. N. Hobson, A. D. Wheatley, Anaerobic Digestion: Modern Theory andPractice, Elsevier Applied Science, London 1993.

[28] V. Yilmaz, G. N. Demirer, Improved anaerobic acidification ofunscreened dairy manure, Environ. Eng. Sci. 2008, 25 (3), 309 – 318.

[29] S. Ghosh, Solid-phase methane fermentation of solid wastes, Trans.ASME 1985, 107, 402 – 405.

[30] E. Rustrian, J. P. Delgenes, N. Bernet, R. Moletta, Simultaneousremoval of carbon, nitrogen and phosphorus from wastewater bycoupling two-step anaerobic digestion with a sequencing batch reac-tor, J. Chem. Technol. Biotechnol. 1998, 73, 421 – 431.

[31] V. Vigneron et al., Nitrate and nitrite injection during municipalsolid waste anaerobic biodegradation, Waste Manage. 2007, 27 (6),778 – 791.

[32] I. Maharaj, Acid-phase anaerobic digestion of primary sludge: therole of hydraulic retention time (HRT), temperature and a starch-rich industrial wastewater, M.Sc. Thesis, The University of Manitoba,Winnipeg, Manitoba 1999.


Low-Strength Wastewater Treatment with Combined Granular Anaerobic and Suspended Aerobic Cultures in Upflow Sludge Blanket Reactors

Documents